Galen recently wrote a wonderful response to my “Fool’s Gold” essay, trying to pinpoint where he and I disagree about when it comes to pragmatic theories of representation. This letter won’t make much sense unless you’ve first read his salvo.

Epistemic status very much “Fumbling around in the dark”—lots of ideas and discourses invoked that I don’t fully understand. Welcoming any commentary or critique; my email is suspendedreason (at) gmail.

*

NB: I’m from the Central Coast of California, which puts me smack-dab on that NorCal—SoCal Mason-Dixon [mentioned in “Fool’s Gold“]. So—rather predictably—I identify as bipartisan and see both views as compatibly, partially true. Which is to say, useful (“carrying true & relevant senses,” “recommending adaptive actions”) across different situations.

I’ll respond as if your post hadn’t ended on a note of reconciliation, the suggestion that our disagreement is mere terminology. In part because it’s more generative; in part because I sometimes suspect that most philosophical disagreements chalk up to “mere” terminology, because of how many possible senses the employed terms carry (“could be taken to recommend”). 

I’ll start by clarifying my stance, on the points you propose under “Agreements.” Then I’ll ask some questions about your stance as outlined in “Disagreements.” And then I’ll push back a bit more aggressively against one common anti-pragmatist stance, which you may or may not sympathize with. 

I’m not very happy with this letter, because it still feels as if so many small root-tendrils have to be sent out, to map the whole space, to give a good compression of the whole, and every time I type, I feel that empty unexplored space expand before me. I tried to ground us a bit with an overarching metaphor of a lock and key.

1.

Your splitting of a weak and strong (or moderate ontological and extreme metaphysical) pragmatism seems like a fair distinction, altho I admit I’m still not totally sure what you mean by ontological. I haven’t read philosophy deeply or dedicatedly enough to position historical thinkers along a moderate-to-extreme axis, but I can confirm that it successfully brackets out a set of beliefs I (we?) don’t endorse from those I (we?) do. 

Some nitpicks:

You write of extreme metaphysical pragmatism:

An agent can cause arbitrary transformations of “reality” via communication and belief…

I don’t love “arbitrary” here; I think few theories endorse an unbounded arbitrary application of their concepts, and so “arbitrary” gets used regularly in these kinds of discussions to mount scarecrows. I think even the most committed “metaphysical pragmatist” would put some bounds on belief’s capacity to transform the world; these bounds are implicit in whatever mechanism(s) this metaphysical pragmatist would ascribe to world-transformation-via-belief. Moreover, my guess is even anti-pragmatist thinkers would concede that communication and belief changes the world all the time. Who can look at the Catholic Church and say otherwise? But a certain rationalistic, scientific person would also want to say: Christianity might be transformative, it might even be culturally adaptive—but it sure isn’t true. And I’m sympathetic to this stance! But part of me then wonders: “Well, what sort of ‘truth’ are you after then? And what are its merits?”

I see extreme metaphysical pragmatism all the time on TikTok: “Bro, you constantly manifest your entire reality; why are you manifesting pain and trauma?” Alfred Adler says: Not so crazy a question! and in the specific context I agree: “trauma” seems to be a viral meme that actively decreases many individuals’ agency and harms their mental health. (While, I’m sure, being tremendously helpful to a subset of people.) But the problem, as you point out, is there isn’t even an attempt at rigorously bounding what can be changed through belief: “your entire reality” is magickally manifested. Nevermind that (at the very least) other individuals are also constantly “magickally manifesting” their reality, and their magickal powers might be stronger than yours, especially in the aggregate. (More eco-logic; the original sin of extreme metaphysical pragmatism may have been solipsism.)

Continuing:

Language and belief do not (and cannot) “represent” the world in a perfectly neutral way…

Another nitpick: I don’t love the use of “perfectly” here. We could also say “in a remotely neutral way.” If we believe that the whole point of representing is to efficaciously act on the environment, then all representation or knowledge is merely a delta of transformation between “us” and “not us” (environment) in service of the “us.” (Note the eerie echoes with inter-group competition.) Representation and knowledge are helplessly, irrevocably, non-neutral. The concept of a “neutral” representation is incoherent. Are we agreed on this? I feel like we’re agreed on this, but maybe your wishy-washy’s doing work. 

*

One metaphor I have liked is to think of a representation as a key. There are billions of locks in the world. For every conceivable key I could create, there is some possible lock which it would open, but that lock may or may not exist in the present world. If I find a key in the street, it is functionally worthless unless I know which lock it opens. An “alien” (i.e. someone not trained in our cultural associations of locks and keys, and in our cultural practices of using keys to open locks) could do nothing with a key; it would be a piece of junk. 

It is true that, for a key to open a lock, there must be some “fit” or structural correspondence between key and lock which “exists” in the Real World separate from belief. But the strange idea of a key “representing” a lock already breaks some of our “picture theory” intuitions about representation. What would it mean to say that a key is “true” except that, in our experience, it opens a given lock? If on a winter’s night, the cold has jammed  up the lock mechanism, is the key no longer “true”?

My inclination, which I readily cede may be wrong, is to think that there’s not much else we can say except that a key does or doesn’t fit a given lock at a given moment, when used by a given individual. There are also altered expectations and altered behaviors, but those are bundled up with the consequences: we can consider, in turn, each facet of the pragmatist crystal, but the crystal is the same.

I also think we can keep improving on the lock and key analogy—continue making it more like linguistic representation—if we use the concept of a key pad. Let us assume that all locks in the world have a 3×4 grid of unlabeled buttons, like a telephone keypad. One conventional system for conceiving of these buttons goes, from top-to-bottom and left-to-right, 1-2-3-4-5-6-7-8-9-*-0-#. But these numbers do not “exist” in any way in or on the keypad itself—they are merely a means people use to refer. 

One quasi-hyperstitional gambit (in the “DJT” sense) works like this: We’re together at a cafe that lacks a restroom. You want to walk to my nearby apartment to use my bathroom; I give you the key-code. You leave the cafe. I phone my roommate, and tell him, “change the locks to accept XYZ key-code.” He goes downstairs and promptly changes the setting. By the time you reach my apartment to use the code, it works. Did I give you a “true” key-code? And does the answer depend on whether my roommate successfully changes the lock before arrive? These are silly terminological questions, but the point is that, if we believe representations are constructed in the service of efficacy, it is difficult to not call the key I hand you an efficacious representation. And I am not sure what “truth” would mean beyond this efficacy. You might want to improve my utterance—might think I was being “more truthful” if I told you, while handing you the key in the cafe, that my roommate would be changing the code for you, or that the code would only unlock the apartment door “in the future” (rather than some theoretical instantaneous “now,” were you able to teleport to my apartment door). But as you well note (“When people say that some claim is true…”), all language brackets a thousand fuzzy assumptions that are expected to generally hold. Must an utterance (or representation) comprehensively describe to be true? If so, no utterance is ever true. What we really seem to care about is robustness or reliability: we want to be told that the roommate is changing the lock if and when we think there is a high probability the swap will not be finished by the time we arrive at the apartment. (So that, if we try to unlock the door and fail, we might make a good decision, such as waiting a few minutes and trying again.) We seem, then, to criticize specifically those gambits we think will be unreliable as promises—which makes sense, because reliability and efficacy are so connected.

*

Some further complications. Let’s say that I abide by the keypad labeling/reference system which goes, from left-hand top to right-hand bottom, 1-2-3-4-5-6-7-8-9-*-0-#. I tell you that the locker keycode is 1-3-5-7-#, and you go off to open the locker. However, you use an alternate reference system, which begins on the symbols * and #, then goes from 0-9. You type in the keycode and the door doesn’t open. The analytic philosopher asks: “Was that a true key-code?” Or, perhaps I try to deceive you telling you the locker keycode is *-0-2-4-9. You go to punch it in and—voila, by a miracle it works! (Since *-0-2-4-9 in your reference system is equivalent to 1-3-5-7-# in his.) The analytic philosopher asks: “Was that a true key-code?”

What the pragmatist—I’ll use William James here, since I think his classic “What is pragmatism?” example of a squirrel going “around” a tree is closely connected to my point—What the pragmatist wants to say is this: The only reason we are concerned with “truth” to begin with is because representations “carry senses” which are varyingly efficacious. Rather than getting stuck on playing conceptual analysis with the word “true,” we should simply note that (1) *-0-2-4-9 was a useful representation for you (2) 1-3-5-7-# was not a useful representation for you (3) 1-3-5-7-# was a useful representation for me (4) *-0-2-4-9 was not a useful representation for you. 

The pragmatist would say: When faced with philosophical disputes, such as whether a key-code is “true,” it is often useful to to reconsider truth as a form of usefulness.

But I think we can add one more complication to better analogize the social representation landscape. Most “intrinsic” situations—interactions between an agent and a non-agent, e.g. you pressing buttons on a mechanical keypad—work as above. But many situations in the world are “extrinsic,” and here, rather than a keypad mechanism, our best analogy is the passphrase. I tell you my apartment’s passphrase is “Billy Budd.” You leave the cafe and walk to my apartment. I call my roommate, and tell him that today’s passphrase is “Billy Buddy.” You reach the apartment door, say “Billy Budd,” and are admitted. 

This is why I am a bit confused when you say:

In day-to-day life—which is to say, when [Suspended] is buying food for his cat or beer for his friends—I do not expect that he is actually confused. He’s a smart guy! He functions in the world!

You set this up like it’s an intrinsic problem, when it isn’t at all! It is one of the most archetypal extrinsic interactions on the planet! Money! Perhaps the most hyperstitional object in our daily lives! It exists purely as promise, an expectation of value upheld through social belief. When that belief collapses, so too does the system of exchange. When I hear that hyperstitional gambits are a form of defecting, I get confused. What is money, except expansion of agency through collective belief? What is humanism, the belief in the “inherent dignity and inalienable rights of mankind,” an idea that (from a progressive perspective) has remade the world incalculably for the better? (Much of your opposition to cybernetic control states, for instance, seems to prefer a humanistic stance, enchanted though it be.) The question is: Do we want to admit these sorts of moves in public language games? And I think we sometimes do.

There may well be useful distinctions to be made between the epistemological status of “money has value” and F=ma. But I do not think that statements of the form F=ma are the only useful or prosocial statements to make. I do not think we are betraying the engineer, and playing zero-sum social games, when we set up a monetary system, or write a text like the American Declaration of Independence, which make hyperstitional gambits or promote aspirational values as if they were simple statements of fact in way F=ma. 

2.

This starts getting to the crux of my questions about your “Disagreement” section. You’re much better-informed than me anthropologically, but I feel skeptical about some of the distinctions you make between modern Western society (cybernetic control state with palace politics and endless extrinsic games) and non-modern, non-Western societies. And yet, when I consider traditional and religious beliefs in indigenous societies, what I see is a cybernetic system for coordinating! Is it really a Western idiosyncrasy for individuals to make themselves “predictable in a way conducive to smooth functioning” in the broader society? It seems to me a cultural universal that those who seriously deviate from predictability sooner or later end up dead or exiled. (That making oneself predictable is usually inextricable from coordinating/cooperating.) What is a medicine man, or elder, or prophet, except one who uses the “thermostatic” leverage (as a trusted, high-information member of the social whole) to alter the behavior of the furnace (his village)? When stories of sky gods and chimeras are told—in order to justify moving the location of the village, or enforcing a certain coming-of-age ritual—are they any more “truthful” in the F=ma sense than humanism or money, or than my telling you the passphrase is “Billy Budd”? And what cultures do not operate on a level of faith and shared belief—belief at the level of “the sacredness of marriage,” or “the sacredness of the totem pole”—which in some conceivable circumstance, or if applied to intrinsic games would prove at best useless, or at worst counterproductive? (Because they are “false” in the sense of being “partial”—bracketing decision-relevant information—or implying senses that are not presently productive. E.g. if you run out of wood and are freezing to death, maybe you should chop down the totem pole for wood. This relates back to the earlier point about robustness and completeness of information—we treat utterances as completely informative when they handle the situation/s we actually encounter.)

A functionalist or evolutionary approach to culture says that by and large, practices which in some way benefit communities persevere. Behaviors and ontologies are not arbitrary but pay rent. And I guess I see an analogy between these sorts of behaviors, and “magical” or enchanting communication. And also an analogy with “inexact” utterances (mythological, literary, poetic accounts of the world), since these crude coarse-grainings cannot be taken too literally, cannot have too much weight put on them, or be taken too fervently as gospel, or they collapse—but if they are taken appropriately lightly, they can be quite useful.

Because crucially, the mechanism “listed on the tin,” for how the behavior pays rent, is not necessarily how it actually pays rent. And that’s because our usual idea of “written on the label” is confused: there is no label, and there are no numbers hard-coded into the figurative keypad. The “label” is whatever learned sense by which fluent cultural members take the belief and translate it into action in certain situations. My interpretation of pragmatism is that the only coherent and meaningful meaning of a representation is how it alters practice. (Which can be varyingly specified or generalized from one individual at one moment, to an entire population of cultural members.) And this the primary mechanism which drives the evolutionary epistemology process by counter-productive utterances and beliefs and practices are weeded—not just on a time-scale of cultural learning and selection, but also individual learning and habitus-selection, so that only functional tactics remain. If this is the case, it may not be coherent to say that (e.g.) a claim of (or belief in) a totem pole’s divinity is true or false. Rather, one can only say that, if an individual’s belief in the sacredness of the totem pole leads him to freeze or starve in winter famine, then it has mislead him—that the sense or behavioral prescription he has taken away from the representation has proven maladaptive. Just as the key only becomes “false” when the lock jams in winter cold. Just as Newtonian physics only becomes “false” when used to solve very specific scientific problems. Is there more we want to designate, when we’re tempted (as I am! but resist) the desire to call the notion itself of a totem pole’s divinity “false” en toto, or try at least to bracket it from F=ma (with the implicit, normative suggestion that F=ma is a preferable form of language production)? Perhaps more to your interests, I guess I’m asking: What makes Western ways of knowing special?

I want to apply this same hermeneutics of suspicion to our own utterances that we call “true.” There is a sense by which we, as fluent cultural practitioners, understand how to change our actions—within certain “types” situations which we, as fluent cultural practitioners, have learned to recognize—in light of some piece of information, in light of some encountered representation of the world. And this seems like the only sense in which “true” could ever meaningfully refer, because it is the only way words seem to “mean.” I think you agree with this when it comes to the idea that models don’t represent on their own, but are part of processes of representing and interpreting. But then—what is the “deeper” level of a “true statement” that you’re after, when you want to sanction the reality-creating beliefs and practices I am more tolerant of in “Fool’s Gold”? 

Perhaps a given society believes that humans have an intrinsic spirit, which can be easily contaminated, and it is important to members’ essential dignity and metaphysical value to keep that spirit pure by avoiding certain acts and associations. Taken metaphysically (evaluated “as written on the tin,” in the language of science) this is nonsense. In practice, this belief leads members to avoid consuming foods with high lead levels, and minimizes transmission of sexual diseases. Is this much different from Dud dreaming up a whacky quest to get these prophetic scrolls in Mexico, which in practice means the group bonds and brings in resources and becomes more stable? In both cases, it’s exactly and only at the point where the belief becomes maladaptive that it becomes, well, maladaptive! And is there anything else worth saying about it? Do the group members really believe in “on the tin”/literal sense of this vision quest—that the alchemical secrets to the philosopher’s stone, which confers biological immortality, are encoded on the scrolls—or is there a sense of playing along that is metaphorical and perfectly consensual, this living in a fiction which makes life easier? 

Can we say that in Lodge 49, when Blaise does take alchemical gospel “in the wrong sense”—so that his obsession with cracking the stone’s secret brings him to the brink of madness—that his interpretation is bad? Similar to the way that a person who uses the keypad reference system of *-#-0-1… (top-left to bottom-right) takes the wrong sense away, when you or I (1-2-3… reference system users) tell her the keycode, and she is unable to open the door? Is literalist language actually, generalizably less ambiguous in communicating the senses by which it should be transformed into action? Or is it just the shared cultural schema we personally are fluent in, and so we’d like to enforce it on others, and call those representations which operate on a different logic of sense-recommendation “false”? These are open questions for me. But I feel an inclination—which I haven’t yet learned to articulately justify—towards sheltering, protecting, incubating in my own life a set of conceptual frameworks which are more varied than literalist language allows. That keeping these conceptualizations alive isn’t just good for me, it’s good for the people around me. Isn’t the most pertinent difference between the engineer and the diplomat merely that the engineer (likely rightly) perceives himself to be in a cooperative game, and the diplomat (again, likely right) perceives himself to be enmeshed in conflict?¹

The desire I see often in the rationalist diaspora (e.g. in some of their attitudes towards high art/literature/culture) is to disparage representations that don’t communicate in literalist language or pay rent in exactly the way they say they do on the tin. I can’t help but wonder if this is a kind of “legibilizing” force, in a James C Scott sense of high modernist control systems. The sort exemplified in contemporary sexual consent cultures of explicit asking & answering. An attempt to build (and normatively enforce) cultural Schelling points of sense-making and sense-taking. To erect an interpretive monoculture of the senses we take away from language. The individual—science’s single unified framework foisted upon him—cannot even opt-in to live inside the logic of myth, or the language of metaphor; any lack of officially auditable transparency—any whiff of manipulation’s possibility—is deemed immoral. If am to be accused of siding with the scheming palace politician, then I accuse my accusers of the very globalism they say they detest!

*

Finally:

It seems as if the basic cybernetic model—the idea that individuals in a society change their behavior based on the communications (sensory inputs) passed to them, and that their changed behavior in turn changes other members’ sensory inputs—is, in the literalist mode of representation, incontestably true. Effectively a tautological or “analytic” description of what it means to be an agent, and what happens when two or more agents interact. But by embracing the cybernetic model, Western states turn a descriptive “fact” into an organizational prescription. My endorsement of a “true” cybernetic model puts me on the side of the bad guys. Which makes me the rogue diplomat whose questioning of official narrative must be shut down, lest it wreak havoc! No? Am I speaking crazy?

Well, it seems to me like many arguments I’ve encountered, while defending strategic & functionalist accounts of language, want to have it both ways—to enforce a taboo against such functionalist accounts as “false” on the grounds that they are harmful. In other words, to exalt the engineer and condemn the diplomat, while also making a diplomatic case against engineering!

Perhaps a case study would help ground some of this out. Say something if this letter doesn’t seem productive. I can’t help but feel that, even as I tried to address your points one by one, I somehow systematically neglected them. 

Yours, 

Sus

1. I’m responding, here, to Galen’s invocation of Ben Hoffman’s “The Engineer & The Diplomat” ↩︎